Managing direct mail campaigns for maximum response

By Pat McGrew

ZIP+4 In Action

Fenske Media, a Rapid City, South Dakota-based printer and direct-marketing solutions provider, offers a web-to-print solution that makes it easy to create and implement solicitation and giving campaigns. Over the last six years, Fenske has watched as response rates began to decline, despite its data acquisition and analysis. To reverse the trend, Fenske deployed a research methodology that used three types of data analysis to identify likely donors for the various fundraising campaigns it runs for clients:

• Factor analysis of the key variables touching socio-geodemographic indicators and integrating those results into the database to embellish the existing data.

• Cluster analysis to zero in on the groups that  actually exist in the database: What is the same about them? Do they share all characteristics or just some? Are they the characteristics you care about? Just because everyone in a cluster lives in an apartment building doesn’t necessarily mean you care about that aspect.

• Discriminant analysis to find the factors that separated responding clusters and non-responding clusters. There were the response drivers used to predict which prospects in which geographies had the highest probability of responding.

All three methods were deployed for each state to allow Fenske to better understand the dynamic nature of that market (factor analysis), the client groups who purchased goods and services in that market (cluster analysis) and the probability that each target in the database would be interested in direct solicitation (discriminant analysis).

That information was used with ZIP+4 postal information to identify new potential donors in a test. Each ZIP+4 postal code was converted to specific coordinates that allowed a broad set of external socio-geodemographic data to be merged with the existing data set. There were three mail drops. The first to test the direction of responses, the second to revise the model and capture more data on factors that impacted conversion and, finally, the third to provide a clearer predictive model.

Response rates for Fenske’s pre-analysis campaigns were under 0.5 percent. Actual test results showed a better than 2 percent response—more than enough to prove the value of the analysis.

Every direct mail campaign—whether it's a solicitation letter seeking new customers, offers for existing customers or a catalog full of products—should be managed for maximum response. However, in the rush to meet deadlines with limited resources, many organizations repeatedly choose to run the same campaigns knowing they're not getting the best response.

Here's how to break those ruts by applying data-driven processes to your campaigns.

The Real Baseline

Start with some rules of engagement for data-driven campaigns. There aren't many, but they are important.

1. Know where you are. In most organizations the assumption is, because there are marketing plans and agreed goals, everyone knows the baseline for response rates on current campaigns. In many cases, that assumption is not supported with fact. To know where you are, you must be able to answer the following questions:

• How many individuals are you trying to communicate with?

• How many individuals have responded in some way to each of the last three campaigns?

• Did each of the last three campaigns grow in response rate?

2. Know where you are trying to go. Oddly, many companies embark on direct mail campaigns without defining an engagement goal. In some companies, the direct mail campaign is even a legacy event—it's done because it has always been done.

To be effective, a direct mail campaign must have a real goal that identifies engagement expectations by customer segment, anticipated response rates, target revenue goals and a means for tracking each of the anticipated data points. The basic questions are:

• What is the revenue goal for the campaign? You might express it as revenue per mail address, revenue per neighborhood or region or revenue by type of customer. But you need a goal.

• What are your target customer profiles? Are you going after your usual suspects, or looking carefully at customer spending patterns? Can you identify your top spenders and your non-spenders? Start with three segments: 1) customers who do not spend regularly or in high enough dollar amounts, 2) customers who spend regularly and in high dollar amounts, and 3) everyone else.

1 2 3 All Next »

0 Comments

View Comments

Companies:
Eastman Kodak Co.

Pat McGrew Author's page

Pat McGrew, M-EDP, CMP is the Director and Evangelist for the Production Workflow Service at InfoTrends. As an analyst and industry educator, McGrew works with InfoTrends customers and its clients to promote workflow effectiveness. She also has a background in data-driven customer communication, and production printing with offset, inkjet, and toner. Co-author of eight industry books, editor of "A Guide to the Electronic Document Body of Knowledge," and regular writer in the industry trade press, McGrew won the 2014 #GirlsWhoPrint Girlie Award for her dedication to education and communication in the industry, and the 2016 Brian Platte Lifetime Achievement Award from Xplor International. Find Pat on Twitter as @PatMcGrew and LinkedIn.

Direct Mail: Finding Lift in Your Data